Added feature for using the tool in a distributed computing environment. (See here for details: http://www.diamondsearch.org/index.php?pages/distributed_computing/)
Fixed an issue that could cause increased memory usage and runtimes in certain cases.
Fixed a bug that could cause a crash when...
Added the sensitivity modes --very-sensitive and --ultra-sensitive. Both modes are designed for finding distant hits of <40% identity with a sensitivity similar to BLAST, with the ultra-sensitive mode being the slightly more sensitive mode.
The --block-size/-b parameter is set to 0.4 and the...
There was another bug, please try again using this commit: https://github.com/bbuchfink/diamond/commit/3f4da374b64be83584cd0c2cada42f2250bf924e
(you have to check out the dev branch)
You will also need to build the db file again.
There seems to be something going seriously wrong. If you got diamond from conda, could you instead try to download the binary from here and/or compile from source: https://github.com/bbuchfink/diamond/releases/tag/v0.9.35
Please double check that you are using the same version for each call to diamond using diamond version. Also use diamond dbinfo -d <db_file> to check the database format version, it should be 3 for the latest version.
Fixed a bug in diamond view that would cause high memory usage and erroneous output.
Reduced the use of temporary disk space.
Fixed a database compatibility issue with big endian architectures.
Fixed a bug that would cause a crash for query sequences shorter than 5 letters in blastx mode.
If you can combine your databases to a larger one, that would be better. Diamond is more efficient the larger the query and database files. That is within certain limits, if your files are a couple of GB in size you can also run it separately.